Praneeth Karimireddy

Research Summary

• Large-scale Private & Federated optimization. Medical data is subject to strict privacy regulations. How can we privately train ML models on data distributed across multiple hospitals without the data leaving the hospitals? How can we share common information across hospitals while personalizing models to the unique aspects of each one? This uses tools like Federated Learning, Differential Privacy, and optimization for large-scale machine learning.
  
• Data Valuation and Data Markets. The data commons that current AI relies on is disappearing. In order to build a sustainable data-ecosystem, people need to be compensated for their data. But, how much is a specific data point worth?. This questions requires understanding i) how data affects uncertainity in a ML model, and ii) the relative importance of datapoints
• Trustworhty AI for Health. Healthcare comes with tons of data challenges on top of privacy concerns: the data may be highly heterogenous, and have missing features. Further, because of the high-stakes involved, fairness and equity, reliable uncertainity quantification, and interepretable predictions are all extremely important.